The current severe accident management measure for Boiling Water Reactors (BWRs) in Japan adopts pre-flooding of the drywell (D/W) floor when the reactor pressure vessel (RPV) is expected to fail to prevent failure of the primary containment vessel (PCV) (wet cavity strategy). This prioritizes rapid cooling of the corium and is expected to lead to more steam generation at the time of the RPV failure. The purpose of this study is to evaluate the influences of changes in the accident management measure on PCV integrity / failure mode and the amount of radioactive material released to the environment for BWRs. The accident scenario has been tentatively defined as the TQUV scenario, in which the water injection is assumed to fail after the RPV depressurization. A set of MELCOR analysis models, which had been used for the Fukushima accident analyses has been adopted to represent TQUV event of BWR4/Mark-I. The effects of fuel coolant interaction (FCI) has not been considered in the current modeling. The analysis results show that in the case without pre-flooding D/W, the D/W wall temperature exceeded 200°C (which may be regarded as a rough estimate for the PCV failure by overheating) in about 30 minutes after the RPV failure. On the other hand, if water is continuously poured onto the D/W floor after the RPV failure, overheating failure of the PCV can be avoided, but over-pressure failure may occur within a few hours after the RPV failure.

According to the current Boiling Water Reactor (BWR) accident management procedure, the Reactor Pressure Vessel (RPV) is not depressurized until the core water level has decreased to a predetermined threshold level (e.g., 10% of the active fuel length from the bottom of the active fuel (BAF)) when the backup low pressure water injection is unavailable. In the meantime, significant Fission Products (FPs) may be released due to core degradation. In such case, excessive FP deposition on the RPV dome may lead to heat up and failure of the top flange sealing of the Primary Containment Vessel (PCV) due to radiation heat. For BWR4/Mark-I TQUV event, the relationship between the different depressurization timings and FP release and depositions (represented by CsI and CsOH) on RPV (especially, steam dryer and RPV dome), D/W, W/W have been highlighted with MELCOR-2.2 analyses. Overall, it may be understood that the later the depressurization timing, the later the RPV failure timing, but the larger the amount of FP released to RPV. The remaining FP not trapped by water pool were mostly deposited on the RPV steam dryer. Clear correlation between the FP deposition amount on the steam dryer and its temperature could be confirmed. However, more investigations may be necessary to reveal what determines the RPV dome temperature, which seemed to have limited influence on the D/W head temperature profile.

The thimble tube, which is made of Zircaly-4, is one of the main components of a PWR fuel assembly. The thimble tube has an important role as a structural member of the skeleton. Another role of the thimble tube is to guide a rod cluster control assembly (RCCA) for insertion during the reactor operation, and the function has to be assured not only in normal operation but in a seismic event. In a horizontal seismic event, the fuel assembly vibrates laterally, which gives bending moment to the thimble tube. In addition, axial compressive force acts on the thimble tube in a vertical seismic event. The integrity of the thimble tube has to be maintained while this force and moment act. Mitsubishi has confirmed by the elastic stress analysis that the stress of the thimble tube is lower than the limit value requested for the seismic event. The stress evaluation method is based on the ASME code. The ASME code also describes the limit analysis which is available when the predicted stress is beyond elastic region of the material. In the analysis, the material is assumed to be elastic-perfectly plastic, and the maximum load that the structure can carry is calculated. For the reason mentioned above, the allowable limit of the thimble tube should be determined as a function between the force and the moment. We are planning to examine the allowable limit experimentally. As a step before testing, an analytical approach for the limit is discussed in this paper. Firstly, the allowable limit is calculated by a beam model assuming elastic-perfectly plastic material, based on the ASME code. Secondly, a 3D model analysis with elastic-plastic material is performed to predict the practical strength. Based on the comparison with the analysis using elastic-perfectly plastic material, ASME based limit is considerably conservative compared with the one with the actual stress-strain curve. Conversely, this means there is enough room to rationalize the allowable limit. As the future work, the experiment will be conducted to obtain the practical limit of the thimble tube and to verify the analysis results.